Causality in public health

Conditions for causality:  A (contraceptive pill) causes B (congenital birth defect) if and only if (1) A is prior to B, (2) change in A is correlated with change in B, and (3) this correlation is not itself the consequence of both A and B being correlated with some prior C.

Sen. Vicente Sotto III has not definitively demonstrated causality.  As far as can be ascertained, there are no documented cases of contraceptives causing birth defects.

It is possible that some other factor C was involved, e.g., other drugs that his wife, Helen Gamboa, may have taken, or chemicals that she may have been exposed to, or genetic factors, or environmental factors.

It also seems that the contraceptive with the brand name “Diane” that senator’s wife allegedly used was introduced in the Philippines way after her pregnancy and birth of her child, so condition 1 above is not met.

Other criteria one looks for to judge whether an association is causal are:

(a) the strength of the association;

(b) dose-response relationship;

(c) consistency of the association across different studies;

(d) specificity of the association;

(e) timing of the cause and the effect—the time-inconsistency problem of the Diane brand of contraceptives used by Gamboa has been scrutinized by the social media, but has not been adequately explained by Sotto; and

f) biological plausibility, or coherence with existing information on the etiology of the disease.

Finally, the case of Vincent Paul Sotto’s congenital disease represents only one undocumented sample and, so far, there are no similar cases to corroborate it.

Thus, no matter how emotionally compelling the senator’s testimony is, it does not measure up to the above epidemiological standards of causality. Picazo, et al.

Reference: Mausner and Kramer, “Epidemiology: An introductory text.”